I think ARM is going to continue to bite into x86 market share significantly.<p>But this article is wrong to right off x86 so easily.<p>Firstly, power consumption. It's right that ARM has lower power draw than x86. The article is wrong by how much, though. Very low power ARM chips draw much, much less than 2-3 watts. These are mostly for embedded systems, though.<p>The 2-3 watts vs 5 watts for ARM vs Atom isn't too significant. The big problem with Atom <i>was</i> that the support systems (memory controller etc) draw ~20 watts. That situation is being improved for netbook systems atm.<p>For sub-netbook systems, Intel is launching it's Moorestown architecture. This is probably still isn't dropping into the Smartphone market in this generation (despite Intel's marketing: <a href="http://arstechnica.com/open-source/news/2010/01/moblin-linux-on-x86-smartphone-intels-small-step-forward.ars" rel="nofollow">http://arstechnica.com/open-source/news/2010/01/moblin-linux...</a>), but should be great for tablets: <a href="http://arstechnica.com/gadgets/news/2010/05/intel-fires-opening-salvo-in-x86-vs-arm-smartphone-wars.ars" rel="nofollow">http://arstechnica.com/gadgets/news/2010/05/intel-fires-open...</a><p>The article also implies that Intel's foundries are a liability. That would be true if there really was useful "competition in the foundry market". Sure, if you want 45nm+ chips produced, there are a number of foundries that can do it. But once you start looking for 32nm foundries they get a lot rarer, and Intel has just announced it's building its new 22nm foundries. That's a whole generation ahead of anyone else in the industry and is a big competitive advantage (Smaller scale in chip foundries means more performance for the same power, or less power for the same performance.)
As I wrote in a similar thread I think Intel and x86 is already dead - the writing is on the wall - Intel just doesn't know it yet. Below is how I believe a key customer has already planned to leave the x86 architecture as nothing but a footnote in their history (alongside the remnants of PowerPC).<p>---<p>Currently Apple relies on Intel for a major component in a key product. Strategically, Apple doesn't like to have to rely on a single source or supplier for key products. Apple will do whatever is possible to remove this reliance.<p>Hence a prediction: within less than 5 years a Mac will be running on an Apple designed ARM processor.<p>How? By slowly, step by step, providing a way towards this.<p>Step 1. Migrate your OS to the new architecture (e.g. iOS already, OS X not far behind) - done<p>Step 2. Migrate your developer base onto developer tools which you control and can easily change the architecture it targets (e.g. Xcode and LLVM) - done<p>Step 3. Provide a space where problematic applications which use other VMs or rely directly on getting too close to the hardware are not welcome (e.g. a Mac App Store) - announced<p>Step 4. Change the marketplace behaviour so that you control how the majority of applications are distributed and can quickly provide updates without user intervention. Such as an App store.<p>Step 5. Release a new Macbook with an ARM processor, absolutely killing on form factor, price and battery performance that Intel cannot compete with. Encourage your Mac App Store developers to flick a switch in Xcode, to recompile and upload their new Universal (x86 & ARM) versions of their Apps to the Mac App Store.<p>Result: you now control the processor direction and application distribution mechanism for a key product and no longer rely upon the whims of Intel.<p>Apple is all about controlling an integrated experience for their customers. Currently Intel is getting in the way of this for the Mac product.
The tagline from <i>Innovator's Dilemma</i> is that <i>well-managed</i> companies can get into trouble when they are being disrupted. The reason is that it's normally good business to get rid of low-margin products and focus your resources on the ones that make the most money. And then someone takes over the low end and expands into the high end.<p>But Intel is not falling for that one. They have made the Atom, a slow, cheap, low-power chip that competes directly with ARM. That's likely a wise move, but Intel now has the problem that low-margin chips are still bad business. They have to have their expensive best-in-industry fabs make low-margin Atoms, when they would much, much rather have them make expensive Xeons.<p>At one point they made a deal to have Atoms manufactured at TSMC, which would have helped a lot with this problem, but apparently that deal didn't work out. Even if it <i>had</i> worked out, the Atom would no longer have the process advantage, and then backwards compatibility would be the <i>only</i> advantage for x86. With Windows becoming less and less relevant, that's a big problem, considering the technical advantages ARM has over x86.<p>So fundamentally, Intel has a problem that CPUs are becoming commoditized, which means they will either have to take much lower margins or retreat to the high end. Both scenarios are unpleasant for them.
I'm not going to write off Intel yet, but they're in a dangerous position. The number one reason Microsoft's stock has been moribund the past ten years is that Linux took over the datacenter. Imagine how much more revenue they'd have if the millions upon millions of X86 servers deployed since 2000 ran Windows. Instead, they're stuck in a saturated, slow-growth monopoly. If non-x86 chips take off in data centers, Intel will be in a similarly bad spot.
From my experience in running Debian on ARM, I can say that the software difference is really non-existence. Any of the normal Debian OSS software works fine.. I did not run into any problems with software not working except for the closed source programs that didn't have ARM packages. I really can't imagine the architecture shift being that big of a problem, as I'm assuming the bulk of the work for ARM with Debian was just recompiling packages.. I had a working XFCE desktop with common packages like Iceweasel (Firefox), OpenOffice.org, GIMP, and so on that was just as easy to set up as x86.
Is ARM's power advantage really that significant for devices larger than the smart phone / tablet form factor? If an Atom-based CPU consumes ~2-3 extra watts but offers marginally better performance and (more importantly) compatibility with an enormous base of existing applications, that doesn't seem like a very compelling argument for switching.
I ran some rough numbers once. I couldn't find exact numbers but estimated for about $1.5 million worth of ARM-based plug servers, you would have close to enough raw CPU power (in terms of FLOPS) compete with the 2003-version of Google.
I think the largest hurdle for ARM to get over is the preponderance of Windows installations with kernels only complied for x86. Linux and OS X (Mach) already run on ARM, and I think that possibly the NT kernel runs on ARM (Windows phone 7 is ARM, right?) I have trouble seeing Microsoft port over Windows proper to ARM until there's a really strong market for it. That being said, perhaps low power consumption ARM devices will provide that market. Perhaps this is another reason that Apple has their own ARM chip - to be at the forefront of the ARM revolution, displacing MSFT?
Great article. As an example, it's fairly obvious Apple is building up it's internal CPU engineering abilities to inevitably put an ARM in a Macbook. But I don't think it will be in the near future due to breaking compatibility with current software, especially those who use SSE instructions. But who knows, maybe they'll create an x86 emulator for the transition like Rosetta. OS XI maybe?
I don't think its clear that ARM would actually be that competitive with x86 at the high end. Many ARM features, like conditional execution, are great for increasing IPC for traditional in-order designs, but make complicated OoO designs more difficult. As you have bigger more featureful chips the relative cost of x86's decoding stage becomes less significant too.<p>All of which isn't to say that the article is wrong and that ARM isn't about to take over the mainstream (I could see it happening but wouldn't bet on it).
I'm a web developer (C# full stack) with 2 years of experience, I have been attracted to ARM development since the moment I studied it in university.<p>As I'll be mostly in technical functions for some time (3 years?) what should I consider when trying an experience in ARM development? What are the pros and cons career wise and technical challenges relatively to continue to do boring CRUD C# applications?
I do wonder, for some of these higher performance devices could Intel / AMD build a version of their chip that was 64-bit only and removed a lot of the legacy instructions / vector attempts? Its not like a recompile / endian shift wouldn't be necessary going from ARM to x86, so it shouldn't add any time to the conversion. You have to account for all the variations of the x86 in a compiler anyway.
The x86 architecture will win out for the same reason it won out decades ago. Most of the world's software was compiled for x86. I have all sorts of software I've been running on my Windows laptop. When I go to buy a new laptop, it'll be another Windows x86 laptop because my software would cost more to replace than the machine upon which it's running.
Does anyone have any data that they can share on the following -- as a company, if I licensed the latest generation ARM processor (e.g., Cortex-A15) and then factored in any additional design and manufacturing costs (e.g., via a foundry), what would be the total cost advantage of using an ARM chip versus a comparable Intel chip (e.g., Atom)? Thanks!
Just wanted to mention...<p><i>(e.g., A4 in a MacBook Air).</i><p>incorrect. Intel Core 2 Duo 1.4 GHz processor in a MacBook Air
<a href="http://www.ifixit.com/Teardown/MacBook-Air-11-Inch-Model-A1370-Teardown/3745/2" rel="nofollow">http://www.ifixit.com/Teardown/MacBook-Air-11-Inch-Model-A13...</a><p>I guess author wanted to mention iPad
Modern CPUs are pretty low-level RISC stuff internally with x86 instruction set layered on top, right? Now, it would be interesting to see what would happen if someone started making desktop-class cpus with the ARM instruction set instead of the legacy x86 instruction set.
In the article it states people are holding onto their PCs longer because they are satisfied with the capabilities of their older machine. I wonder if this is really proven or a misread caused by the XP -> Vista transition problems.
That will surely happen. CPUs have for some time been limited by heating problems and the architecture with more performance with the same power consumption will win. If ARM can deliver that, then there's no question about it.
Intel is aware of this. Atom is the shot across the bow. They are working on making smaller and lower-power parts. It's interesting to watch ARM race up as Intel races down.
while this might be true for the mobile world, the author does not seem to consider that people do more of their computations in the cloud or data centers that in turn have to purchase great numbers of classical high-performance chips. While people are happy with the performance they get on their home machines the demand for greater computing power in the cloud will grow further.